Looking for a job

Sneha Kiran Manded

@snehakiranmanded

Message

AI-focused Full Stack Engineer building low-latency voice AI for healthcare and production systems.

India

Message

What I'm looking for

I want to build production-grade voice AI and LLM systems in healthcare, focusing on low-latency real-time pipelines, strong API design, and measurable outcomes. I thrive in fast iteration, clear ownership, and systems that scale reliably.

I’m an AI-focused Full Stack Engineer specializing in voice AI systems, LLM-powered products, and healthcare technology. I’ve built production-grade platforms for real patient triage, live dental voice calls, and clinical documentation at startup scale.

Across the full AI stack, I design STT→LLM→TTS pipelines and real-time WebSocket audio streaming. I orchestrate LLM workflows with Gemini (including custom prompt engineering) and integrate REST APIs that handle concurrent load reliably.

I consistently improve performance and delivery speed in fast-paced environments—reducing latency by 35–42% and shipping high-impact features with measurable uptime. At Vocab AI, I built multi-agent planner–executor pipelines (92% task accuracy) serving 10K+ daily requests with 99.5% uptime.

My work is centered on practical outcomes for healthcare users: I’ve shipped a WhatsApp AI triage bot handling 50–200+ patient sessions with end-to-end symptom intake and department mapping. I also built an OT notes speech-to-structured-document pipeline using Whisper STT, plus an AI consultation note-taker that cuts post-consultation documentation effort by 60%—so clinicians spend less time typing and more time caring.

Experience

Work history, roles, and key accomplishments

Current

AI Engineer & Full Stack Developer

Current

Medora Health

Nov 2025 - Present (8 months)

Designed and shipped a WhatsApp AI triage bot in production handling 50–200+ patient sessions, including Gemini LLM-based department mapping and multi-service routing (doctor, lab, pharmacy). Built an RBAC-based multi-role EHR platform and a Whisper STT speech-to-structured-document pipeline for OT notes, reducing surgical documentation effort by 60% while ensuring audit trails and digital signing

Whisper STT RBAC MongoDB

Current

Software Developer

Current

Vocab AI

May 2025 - Present (1 year 2 months)

Developed an audio transcription + NLP scoring pipeline for CRM call auditing, auditing 100+ agent calls and significantly reducing manual QA review time. Designed REST APIs (React.js + Flask + PostgreSQL) and built planner–executor multi-agent pipelines achieving 92% task accuracy, 10K+ daily requests, 99.5% uptime, and 35% lower API latency.

Audio Transcription REST APIs React Flask PostgreSQL Multi Agent Systems Query Optimization Uptime & Latency Optimization

Research & Development Intern

Humanoid.AI Lab (IIIT Dharwad)

Jan 2025 - Apr 2025 (3 months)

Built RoboGuide using Whisper STT and Gemini LLM-based command understanding to convert voice instructions into autonomous robot navigation actions. Implemented a ROS + Gazebo simulation pipeline with intent-to-action mapping, improving simulation accuracy by 30% and reducing testing cycles by 25%.

Whisper STT Speech Command Understanding ROS Gazebo Robotics Navigation Simulation Pipelines

Full Stack Developer Trainee

VTU VRIF

Oct 2024 - Dec 2024 (2 months)

Built MERN-based applications to support research workflows and internal tools. Developed reusable React components for data-heavy use cases and improved UI/UX through iterative testing.

React MongoDB MERN